Objectives: To better understand the effect of admixture on long range linkage disequilibrium (LD), we characterized extended LD in gene-rich regions of an African-American population. Methods: Approximately 290 cM of chromosomes 1, 3, 6, 11–17, 20 and 22 were scanned using 109 polymorphic microsatellite markers spaced an average of 3 cM apart. Disequilibrium between loci (D′) was based on maximum-likelihood estimates of haplotype frequencies computed for 200 unrelated African Americans. Results: Mean D′ values were highest on chromosomes 6p23–p21.3 (D′ = 0.33) and 15p22.2–p25.3 (D′ = 0.34), and lowest on chromosome 12p11.2–q14 (D′ = 0.21). Overall, the variance in LD among chromosomes accounted for approximately two-thirds of the total LD variance. Of the 434 locus pairs spaced between 0.3 and 38.7 cM apart, there was no detectable correlation between LD and recombination distance and a weak negative correlation between LD and physical distance (r_s = –0.12; p = 0.031). For the 192 intrachromosomal locus pairs where allele frequency data were available from the Centre d’Etude du Polymorphisme humain (CEPH), we found a statistically significant positive correlation between LD and the allelic frequency differences (δ) between the African-American study population and Caucasian reference CEPH population (r_s = 0.53; p < 0.0001). The correlation between LD and both recombination and physical distance was markedly increased for locus pairs with high δ levels. Conclusions: Our results suggest that recent Caucasian admixture maintains a high level of long range LD in African Americans on a genomic scale, and selected markers with large African American/Caucasian δ levels may be useful in association studies.

Linkage disequilibrium (LD) is the phenomenon where alleles at two separate genetic loci are found more often together at the population level than would be expected based on their individual allelic frequencies. Possible causes of LD include recent mutation, founder effects, or selection. LD that arises because of these factors is rare and generally observed only between very tightly linked loci. Another potential cause of LD is population admixture. Admixture linkage disequilibrium (ALD) arises when two populations that have been separated for a long time recombine. If allele frequencies at the same loci are markedly different, then LD can even occur between alleles at locus pairs on different chromosomes. ALD between unlinked loci quickly dissipates over a couple of generations, but LD between linked loci decays more slowly and may be detectable 10 to 20 generations after the initial admixture (1). An important concept related to ALD is the method by which the two populations are mixing. Recently, Pfaff et al. (2) described how different models of admixture can affect the observed LD over time. Figure 1 describes the two models. In the hybrid-isolation (HI) model, admixture occurs in a single generation and is followed by recombination and drift with no further contribution from either parent population. In the continuous-gene-flow (CGF) model, admixture occurs at a steady but reduced rate in every generation such that the cumulative admixture at generation G t is equal to the HI model. Whereas the HI model introduces a higher level of LD into the hybrid population in the initial generations, the CGF model retains LD in the hybrid population longer and at a higher level than the HI model after approximately 15 generations from the initial admixture.In the case of the African-American gene pool, the CGF model is probably most appropriate.